Project Summary ? Overall The outcomes of patients with metastatic melanoma have improved dramatically over the last decade due to an improved understanding of the molecular drivers of this disease. In particular, multiple targeted therapy regimens have been approved for patients with a BRAFV600E/K mutation, which are present in ~50% of cutaneous melanomas. These treatments achieve clinical responses in ~80% of patients with a BRAFV600E/K mutation, thus providing proof- of-concept of the therapeutic potential for personalized therapeutic strategies. However, most of the patients will progress within 2 years of starting those therapies. Further, currently there are no targeted therapies that have been shown to be effective in patients with a wild-type BRAF. Thus, there are unmet clinical needs to develop treatments that prevent or overcome resistance to existing therapies for patients with a BRAFV600E/K mutation, and that are effective in patients without a BRAF mutation. In order to facilitate the development of new therapeutic strategies, over the last 5 years we have led a major effort to develop a broad collection of PDX models to reflect the clinical, histological, and genetic heterogeneity of this disease. Our collection of >450 PDX models represents one of the largest collections for any human malignancy, and our initial testing demonstrates that the collection accurately recapituluates the oncogenic drivers and molecular heterogeneity that is observed in patients. This collection also includes a subset of PDX established from patients with acquired resistance to targeted therapies that have been maintained on those agents in vivo to sustain their resistant phenotype. Together these efforts have generated a robust resource to develop, refine, and prioritize new personalized combinatorial therapies for patients. Thus, we propose to establish a multi-disciplinary and multi-institutional PDTC Program focused on the use and continued expansion of our robutst melanoma PDX collection to identify new therapeutic approaches that fill important clinical gaps in this disease. Project 1 will characterize kinase inhibitors targeting receptor tyrosine kinases (RTKs) and the PI3K-AKT pathway in combination with approved targeted therapies in PDX models in treatment-nave and drug- resistant BRAFV600E/K-mutant melanomas. Project 2 will characterize targeting the apoptotic machinery of cancer cells as a combinatorial approach with MAPK pathway inhibitors in PDX with and without BRAFV600E/K mutations, including in subsets with aberrancies in anti-apoptotic genes. Both Projects will utilize baseline characteristics of the PDX models to identify potential biomarkers of sensitivity and resistance to guide personalized patient selection for future clinical testing. In addition, tumors will be analyzed for treatment-induced effects that correlate with sensitivity to identified resistance mechanisms and potential new combinatorial strategies. Finally, markers identified in the preclinical models will be compared to parallel biomarker studies of metastatic melanoma patients enrolled in clinical trials utilizing the same agents. These studies will be supported by the PDX, Bioinformatics, and Administrative Cores to ensure their success, along with a Pilot Project program to open new areas of investigation. Together these studies will have high potential to translate into new personalized clinical trials for patients. 1